Adapter Generation for Extracting and Querying Data from Web

Accessing and integrating data from heterogeneous sources has become a significant challenge. So-called adapters provide the functionality for translating SQL queries into queries understandable by the source as well as converting the results into a common model. In this paper, we present our approach of an adapter for Web sources, which is configurable by specifying a sourcespecific extraction function. We focus on two main tasks: query modification in order to extend the source capabilities and data extraction. The extraction step bases on an operational description, that enables an interactive exploration of the result format during the development phase. Finally, we present our ideas for semi-automatic discovery of extraction patterns by analyzing example documents.

• Type: Conference
• Year: 1999
• Where: WEBDB
• Authors: Kai-Uwe Sattler, Michael Höding
• Keywords: Database, Extraction Step Bases, Heterogeneous Sources, Sourcespecific Extraction Function, WEBDB 1999